Copyright Ó 2006 by the Genetics Society of America DOI: 10.1534/genetics.106.061705

Note

An Efficient Genetic Screen in Drosophila to Identify Nuclear-Encoded With Mitochondrial Function

T. S. Vivian Liao,*,1 Gerald B. Call,†,1 Preeta Guptan,† Albert Cespedes,† Jamie Marshall,† Kevin Yackle,† Edward Owusu-Ansah,† Sudip Mandal,† Q. Angela Fang,† Gelsey L. Goodstein,† William Kim† and Utpal Banerjee*,†,‡,2 *Molecular Biology Institute, †Department of Molecular, Cell and Developmental Biology and ‡Department of Biological Chemistry, University of California, Los Angeles, California 90095 Manuscript received June 8, 2006 Accepted for publication July 6, 2006

ABSTRACT We conducted a screen for glossy-eye flies that fail to incorporate BrdU in the third larval instar eye disc but exhibit normal neuronal differentiation and isolated 23 complementation groups of mutants. These same phenotypes were previously seen in mutants for cytochrome c oxidase subunit Va. We have molecularly characterized six complementation groups and, surprisingly, each encodes a mitochondrial . Therefore, we believe our screen to be an efficient method for identifying genes with mitochondrial function.

ITOCHONDRIAL function is essential for a (Flores et al. 1998). However, in the case of tend, such M number of important cellular processes, such defects are caused by the lack of a sufficient number of as the generation of ATP (reviewed in Ackerman and precursor cells from which cone and pigment cells arise Tzagoloff 2005), apoptosis (reviewed in Newmeyer (Mandal et al. 2005). and Ferguson-Miller 2003), and the regulation of The events that give rise to the tend adult eye phe- aging (reviewed in Ohta 2003). Recently, work from notype occur in larval eye development. The larval our laboratory has shown that the metabolic status of a Drosophila eye imaginal disc undergoes two distinct cell, controlled by mitochondrial function, also regu- phases of proliferation (Wolff and Ready 1991). andal lates the G1–S checkpoint in mitosis (M et al. During the early larval stages prior to differentiation, 2005). This is evident in tenured (tend) mutants, which all cells divide frequently and asynchronously, appar- contain a null mutation in the encoding cyto- ently without regulation. In the third larval instar, the chrome c oxidase subunit Va of complex IV of the morphogenetic furrow sweeps across the eye disc to mitochondrial electron transport chain. This mutation pattern the cells (Ready et al. 1976). Within the mor- causes a reduction in ATP generated in mutant cells to phogenetic furrow, cells arrest in G1 and, following the 40% of wild-type levels, triggering the activation of passage of the furrow, either differentiate into neuronal AMPK and p53, which leads to the eventual down- photoreceptors or reenter the cell cycle and undergo a regulation of cyclin E. A G1–S mitotic checkpoint is terminal round of division termed the second mitotic then enforced, preventing cells from reentering the wave (SMW). This generates the precursors for cone cell cycle (Mandal et al. 2005). and pigment cell lineages (Wolff and Ready 1991). Adult tend mutants have glossy eyes lacking lens Misregulation of the SMW will lead to a loss of these material, normally secreted by cone and pigment cells accessory cell types. In the third larval instar, cells of the eye. In earlier studies, the lens-depleted glossy-eye mutant for tend anterior to the furrow slow down in cell phenotype was described in lozenge mutants, which are cycle progression, while those posterior to the furrow defective in cone and pigment cell specification fail to cross the G1–S checkpoint due to impaired ATP production (Mandal et al. 2005). Therefore, in addi- tion to undergoing fewer rounds of mitosis prior to the 1 These authors contributed equally to this work. passage of the furrow, tend mutants lack the SMW 2Corresponding author: Department of Molecular, Cell and Develop- mental Biology, 2204 Life Science, 621 Charles E. Young Dr. South, and, as a result, have a dramatically reduced number Los Angeles, CA 90095. E-mail: [email protected] of accessory cells. Importantly, in tend mutant cells, cell

Genetics 174: 525–533 (September 2006) 526 T. S. Vivian Liao et al.

Figure 1.—Crossing scheme for genetic screen. A total of 75,000 were mu- tagenized and screened for the adult glossy-eye phenotype. A total of 90 glossy-eye mutants were isolated, and larval eye-disc clones were induced to study their BrdU incorporation and ELAV expression. Twenty-three failed to incorporate BrdU but expressed ELAV normally and were studied further. m*repre- sents the EMS-induced muta- tion. The cl-R3 mutation is cell lethal and is used to eliminate cells that are homozygous for the that does not contain m*. Similarly, for larval clones, cells homozygous for the RpS3 (Minute)mutation are eliminated. Cells heterozy- gous for the cl-R3 and RpS3 mutations grow more slowly, allowing m*/m* cells to form large clones. divisions in second and early third instar eye discs are chromosomes under the control of the eyeless enhancer, not affected. As a result, tend mutants give rise to large specifically in the embryonic anlagen of the developing clones in the adult eye. This is different from the eye (Newsome et al. 2000). The resulting clones of cells phenotype resulting from mutant clones in the basic will be homozygous for the newly induced mutation and cell cycle machinery, which would not be able to give rise are white due to the lack of the w1 gene essential for to any mutant clones. Furthermore, early patterning pigmentation. A total of 75,000 adult flies were screened and differentiation of neurons are also not affected in for the mosaic glossy-eye phenotype. Ninety mutants tend, as mutant clones express pan-neuronal, as well as were isolated. Complementation analysis was then per- cell-type-specific, markers and extend axons normally to formed by crossing each of the 90 mutants with one the optic lobe (Mandal et al. 2005). another and looking for lethality in the offspring. This On the basis of the nature of the tend phenotype, we yielded a total of 46 complementation groups. All hypothesized that a simple screen that identifies glossy- mutants were balanced over a TM6B Tb, y1 chromosome eye flies, followed by a secondary screen for a subset of and maintained as stocks. the mutants in which cells exhibit proliferation defects Next, third larval instar eye imaginal discs from the but have relatively normal patterning in the third larval identified glossy-eye mutants were assayed for BrdU instar, will potentially identify nuclear genes that en- incorporation and stained with the neuron-specific code mitochondrial . Our hypothesis is sup- ELAV antibody (Yao et al. 1993). Larval eye-disc clones ported by our earlier finding that, like tend, flies mutant were generated by crossing the mutant stocks to yw in the mitochondrial components pdsw, mRpL4, and ey-flp; FRT82B Ubi-GFP RpS3/TM6B Tb, y1 flies. Clones mRpL17 exhibit a glossy-eye phenotype and fail to in- homozygous for the mutation are negatively marked corporate BrdU in the larval eye disc but stain normally by their lack of GFP. The 46 complementation groups with the ELAV antibody (Mandal et al. 2005). In this isolated from the glossy-eye screen fall into four catego- article, we show that an unbiased screen for glossy-eye ries on the basis of this characterization. Ten mutant mutants that exhibit these larval phenotypes is remark- groups exhibit both wild-type neuronal patterning and ably effective in identifying genes with mitochondrial BrdU incorporation, indicating that the glossy-eye phe- function. notype is due to mutations in genes that function after We conducted an eyeless-flp/FRT-based mitotic recom- the third instar and are required for cone and pigment bination screen (Newsome et al. 2000) on chromosome cell specification. Three mutant groups display abnor- arm 3R (Figure 1). y w ey-flp; FRT 82B males were mu- mal ELAV staining patterns while cells incorporate tagenized with 25 mm ethyl methanesulfonate (EMS) BrdU normally. Since the formation of cone cells re- and crossed to y w ey-flp; FRT82B P[w1] cl-R3/TM6B Tb, quires a combination of signals from differentiated y1 females to generate adult eye clones. ey-flp drives neuronal photoreceptors (Flores et al. 2000), this phe- mitotic recombination between the FRT-containing notype likely occurs because of defects in the patterning Note 527

TABLE 1

Mapping results and mutations in nuclear-encoded genes with mitochondrial function that exhibit a G1–S block in the third larval instar with normal neuronal differentiation

A. Mapping results

Cytological position Gene name Mitochondrial process Alleles B. Mitochondrial genes identified by the glossy eye–BrdU screen 84E4 CG10092 arginyl tRNA –tRNA JR70, JR77, JV83, JR101, JR203, ligase (atl) synthesis JV219, JR247, JV250, JP782 84F11 CG9613 coenzyme Q biosynthesis Oxidative phosphorylation–complex III JV259, JP768 protein 2 (coq2) 91E4 glutamyl-tRNA amidotransferase Protein biosynthesis–tRNA synthesis JR15, JV87, JR94, JR113, JR205 A (gatA) 91F1 ATP synthase subunit d Oxidative phosphorylation–complex V JR92, JV115, JR238 (ATPsyn-d) of electron transport chain 94A1 CG6022 cytochrome c Oxidative phosphorylation–complex IV JV32, JV56, JV193, JR214; heme lyase (cchl) of electron transport chain Exelixis c04553 (Thibault et al. 2004) 94B6 mitochondrial biosynthesis– JV28 protein large subunit assembly 45 (mRpL45)

Previously identified mitochondrial genes that cause the glossy-eye–defective BrdU incorporation phenotype 23F3 pdsw Oxidative phosphorylation–complex I pdswk10101 (Spradling of electron transport chain et al. 1999) 35F11 mitochondrial ribosomal protein large Protein biosynthesis–ribosome mRpL4k14608 (Spradling subunit 4 (mRpL4) assembly et al. 1999) 61B3 mitochondrial ribosomal protein large Protein biosynthesis–ribosome mRpL17KG06809 (Bloomington subunit 17 (mRpL17) assembly Drosophila Stock Center at Indiana University) 86F9 cytochrome c oxidase subunit Va (CoVa) Oxidative phosphorylation–complex IV JP785 (Mandal et al. 2005) of electron transport chain

Cytological No. of genes Cytological No. of genes Allele position in region Allele position in region Other mapping results JV282 85A6-B1 24 JV212 98E1-E3 6 JR107 89A5-A8 29 JV47 98F5 9 JV166 96A21-B4 34 JR8 100B9 5 In A, dark shading indicates complementation groups that have mapped to a deficiency and light shading indicates gaps within the available deficiency kit. The candidate regions and potential mitochondrial genes for complementation groups that have not been cloned are also indicated.

of specific photoreceptors. Ten mutant groups exhibit the furrow. They were further analyzed as part of this defective ELAV staining patterns in early clusters as well study. as defects in BrdU incorporation, which can be attrib- We have mapped all mutants from this last category to uted to mutations in genes required for specification of either available deficiencies on chromosome 3R from multiple cell types in the eye. Finally, 23 mutant groups the Bloomington and Exelexis collections (Thibault fail to incorporate BrdU properly but show essentially et al. 2004) or attributed their map positions to gaps normal patterning of neuronal clusters emerging from within the deficiency kit (Table 1A). We have cloned 528 T. S. Vivian Liao et al. six mutant complementation groups, consisting of 24 fact that mitochondrial heme lyases lack an N-terminal alleles, and obtained sequence information on the mo- targeting signal and instead are directed to the mito- lecular lesions of 15 of the alleles (Figure 2). As the chondrion by an internal localization sequence in the guiding hypothesis of this screen was to identify mito- third quarter of the protein (Diekert et al. 1999). This is chondrial mutations, we sequenced potentially mito- further supported by the fact that MitoProt also predicts chondrial genes within candidate regions as determined a very low (4.8%) chance of mitochondrial localization through BLAST analysis (Altschul et al. 1990) and for the S. cerevisiae CCHL. MitoProt scores, which predict potential mitochondrial ATPsyn-d: Studies in Escherichia coli show that F1F0- localization on the basis of the presence of a mitochon- ATP synthase, complex V of the electron transport drial localization sequence (Claros and Vincens 1996). chain, generates ATP through cooperation between a Using this method, we identified four mutants in ‘‘rotor’’ and a ‘‘stator’’ (reviewed in Boyer 1997; Weber CG6022, a homolog of cytochrome c heme lyase (cchl); two et al. 2004). The stator consists of two b-subunits and one mutant alleles of CG9613, a homolog of coenzyme Q d-subunit (reviewed in Dunn et al. 2000), and loss of the biosynthesis protein 2 (coq2); nine mutants in CG10092,a d-binding site on the a-subunit of F1 results in impaired homolog of arginyl tRNA ligase (atl); three mutant growth and ATP synthesis in vivo (Weber et al. 2004). alleles of ATP synthase-d subunit (ATPsyn-d); five mutants Three alleles of ATPsyn-d were isolated from the screen in glutamyl-tRNA amidotransferase A (gatA); and one mu- (Figure 2). JV115 contains a G . A transition that results tant allele of mitochondrial ribosomal protein large subunit in a premature at W54, while JR92 has a C . 45 (mRpL45). Combined with previously described mu- T transition that prematurely terminates the protein at tants in pdsw, mRpL4, mRpL17, and tend, all mutations Q83. A molecular lesion for the third allele, JR238, has exhibiting the glossy-eye phenotype with defective BrdU not been found; presumably, it is located within a reg- incorporation and wild-type neuronal patterning cloned ulatory region of the gene. The mutations all lie within thus far could be rationalized as having roles in the the ATP synthase-d domain of the protein. MitoProt (see below and Figure 3 and Table 1B). analysis suggests that Drosophila ATPsyn-d has a 92% Note that this screen identified the first mutant alleles likelihood of localizing to the mitochondria. for CoVa, coq2, atl, ATPsyn-d, gatA, and mRpL45. Single COQ2: coq2 encodes the enzyme parahydroxyben- alleles of cchl, pdsw, mRpL4, and mRpL17 had been zoate (PHB):polyprenyltransferase (Ashby et al. 1992). identified as lethal mutations in either the Exelixis col- As part of a multi-step process involved in ubiquinone lection or the Berkeley Drosophila Genome Gene biosynthesis, PHB:polyprenyltransferase catalyzes the Disruption Project, but were not previously genetically transfer of a polyprenyl group from polyprenyl diphos- characterized (Spradling et al. 1999; Thibault et al. phate to parahydroxybenzoate to form 3-polyprenyl-4- 2004) (Table 1B). hydroxybenzoate (Winrow and Rudney 1969). Once CCHL: Within the electron transport chain, CCHL synthesized, ubiquinone functions as an electron trans- participates in the function of complex IV by attaching porter between complexes I, II, and III of the electron a heme group to apocytochrome c, thereby converting transport chain (Green 1966). Two alleles of coq2 were it to a functional holocytochrome c and enabling its identified from the screen (Figure 2). The first allele, import through the inner membrane of the mitochon- JP768, contains a 94-bp deletion mutation starting at drion (reviewed in Kranz et al. 1998). In Saccharomyces A100 that causes a frameshift in the remaining se- cerevisiae, strains lacking or mutant in CCHL have quence. All amino acids downstream of the deletion increased levels of apocytochrome c in the are altered, and a premature stop codon is introduced (Dumont et al. 1991). From the screen, we have isolated after the 232nd amino acid. JV259 has a T . A trans- four alleles of cchl (Figure 2) on the basis of their failure version that results in a premature stop codon at L189. to complement Exelixis c04553 (Thibault et al. 2004), Both mutations fall within the UbiA prenyltransferase which harbors a lethal insertion in cchl. The first allele, domain of COQ2. Drosophila COQ2 has 45% identity JR214, harbors a T . C transition resulting in the amino to the S. cerevisiae homolog and a 97% likelihood of acid substitution M25T. JV56 contains a C . T transi- being imported into the mitochondrion according to tion, resulting in a premature stop codon at Q65. The MitoProt analysis. remaining two alleles, JV193 and JV32, have G . A GatA: Aminoacylation studies have shown that there transitions at the first and last of intron 2, is no detectable glutaminyl–tRNA synthase activity in respectively, and thus may result in the insertion of 19 cyanobacteria, Gram-positive , and amino acids if splicing cannot occur normally. All of and mitochondria of plants and animals (Wilcox and the mutations occur within the heme lyase domain of Nirenberg 1968; Schon et al. 1988). Instead, these the protein. Drosophila CCHL bears 53% identity to the organisms and organelles synthesize glutamine by mis- human homolog and 42% identity to the S. cerevisiae charging a tRNAGln with glutamate and then using homolog. According to MitoProt predictions, CCHL glutamyl–tRNA amidotransferases to convert the gluta- has a 28% likelihood of localizing to the mitochon- mate to glutamine (Schon et al. 1988). According to drion; this relatively low likelihood may be due to the MitoProt predictions, Drosophila GatA has a relatively Note 529

Figure 2.—Map positions and molecular lesions of identified mutations. Deficiency stock names and their approximate end- points, as annotated by FlyBase, are indicated. Solid black lines, alleles identified from the screen; red stars, point mutations; brackets, deletions; red bars, deficiencies that fail to complement the mutant; green bars, deficiencies that complement the mutant. low probability (56%) of localizing to the mitochon- The Drosophila genome has three genes encoding drion. To determine the exact subcellular location of proteins with potential arginyl tRNA ligase activity on GatA, we constructed a GatA–GFP fusion protein, the basis of annotation in FlyBase transfected Drosophila S2 cells with the construct, and (http://www.flybase.org): atl, Aats-arg, and CG8097 co-immunostained the cells with a GFP antibody and (Drysdale and Crosby 2005). BLAST analysis shows MitoTracker Orange CM-H2TMRos, a dye that localizes that ATL is 36% identical to Msr1p, while AATS-arg and to the mitochondrion in response to its membrane CG8097 are 28 and 24% identical, respectively. Accord- potential. This colocalization assay shows that GatA is ing to MitoProt analysis, ATL is only 8.4% likely to indeed a mitochondrial protein (Figure 4, A–C). Thus, localize to the mitochondrion. This low number did not both functional and localization studies and the nature support a mitochondrial function for ATL. To pinpoint of other mutants isolated by our screen establish that its subcellular location, we constructed an ATL–GFP GatA has a mitochondrial function. Five alleles of gatA fusion protein and demonstrated that it colocalizes with were isolated from our screen; we have located molec- MitoTracker Orange CM-H2TMRos to the mitochon- ular lesions in four alleles (Figure 2). JR94 has a 168-bp drion by immunostaining S2 cells transfected with the deletion in exon 5. JR15 harbors a G . T transversion, fusion protein (Figure 4, D–F). Thus, ATL’s subcellular resulting in the amino acid substitution Q54H. JR113 localization, similarity to Msr1p, and its suite of phe- has a C . G transversion that causes the substitution notypic similarities to other known mitochondrial mu- H278D. Finally, JV87 has three point mutations, G . T, tants, indicate that this protein functions in the T . G, and G . A, which result in the following amino mitochondrion. Our screen has identified nine alleles acid changes: Q53H, A58S, and V74I. All of the mutations of atl, and we have located molecular lesions in two of lie within the putative amidase domain of the protein. these nine alleles (Figure 2). JR70 has a G . A transition ATL: It has been determined that S. cerevisiae has two resulting in the amino acid change G182R. JV219 con- genes encoding arginyl tRNA ligase (ATL), with one tains a 12-bp in-frame deletion starting at V266. Both functioning in the cytoplasm and the other, MSR1,in mutations occur within the catalytic arginyl tRNA core the mitochondrion (Tzagoloff and Shtanko 1995). domain. 530 T. S. Vivian Liao et al.

Figure 3.—Representative adult and larval phenotypes of mutants from mapped comple- mentation groups. (A–N) Eyes of adult flies ex- amined by scanning electron microscopy (A–G) and bright-field microscopy (H–N). An eye con- taining mock clones is mosaic in color (H), but the entire eye is faceted as in wild-type tissue (A). In eyes mutant for the indicated mitochon- drial genes, cells heterozygous for the mutation are faceted (B–G) and red (I–N) as in mock clones, while homozygous clones are glossy and white. In all images, the representative allele is in- dicated in parentheses and posterior is to the left. (O–U) 30-min BrdU incorporation (red) in third instar larval eye discs with mock (O) and mutant (P–U) clones. Armadillo (blue) marks the mor- phogenetic furrow. In an eye disc with mock clones (O), both green and nongreen tissue are wild type. BrdU is randomly incorporated ante- rior to the furrow (‘‘a’’); posterior to the furrow, a single band of BrdU incorporation (arrowhead) marks the SMW. In mutant eye discs, BrdU incor- poration is lost in homozygous mutant tissue, marked by the lack of GFP, both anterior and pos- terior to the furrow. Incorporation remains nor- mal in clones heterozygous for the mutation (green). For comments on the apparent local nonautonomy of this phenotype (arrow), see Mandal et al. (2005). (V–B9) ELAV antibody stainings of eye discs with mock (V) and mutant (W–B9) clones. Expression of ELAV is largely nor- mal in all mutant clones. Note 531

Figure 4.—GatA and ATL localize to the mitochondrion. Co-immunostaining of GatA–GFP and ATL–GFP fusion pro- teins (A and D, green), respectively, with MitoTracker Orange CM-H2TMRos (Mo- lecular Probes, Eugene, OR) (B and E) shows mitochondrial localization of the fusion proteins (C and F, yellow). Full- length cDNAs were cloned using the Gateway cloning system (Invitrogen, San Diego) into the pAWG vector containing the actin promoter and sequence coding for a C-terminal GFP protein. Blue, DAPI staining (C) and TOPRO3 (F) staining.

mRpL45: mRpL45 is a protein of 599 amino acids tants discussed in this study—tend, pdsw, coq2, cchl, and with a molecular weight of 26 kDa (reviewed in Graack ATP syn-d—function either within or between com- and Wittmann-Liebold 1998). At least 50 mitochon- plexes of the electron transport chain. These mutants drial ribosomal proteins exist within the S. cerevisiae likely activate the G1–S checkpoint as described for tend , among which at least 39 are (Mandal et al. 2005). The remaining mutants—atl, classified as large subunit proteins. In humans, these gatA, mRpL4, mRpL17, and mRpL45—function in diverse ribosomal subunits help synthesize 13 mitochondrial roles within the mitochondrial protein biosynthesis path- proteins essential for oxidative phosphorylation, and ways. The mitochondrial genome encodes 13 proteins mutations in the genes encoding them are associated that are essential for oxidative phosphorylation, and with many disorders (reviewed in Sylvester et al. 2004). defects within the protein system are likely to According to MitoProt, mRpL45 has an 88% chance of cause downstream effects in energy production and localizing to the mitochondrion. One mutant allele of subsequently affect the G1–S transition of the cell. mRpL45 was identified from the screen (Figure 2). JV28 The screen described here is the most efficient in any harbors a 9-bp deletion and a 7-bp insertion in exon 2 higher for identifying nuclear-encoded genes that results in a frameshift starting at V230 and a that function in the mitochondrion. All six mutants premature stop codon at the 279th amino acid. The mapped thus far are likely to have roles in mitochon- mutation occurs within the TIM44 domain, which is drial processes. This is further supported by the fact that involved in transporting the protein across the inner previously described P-element mutants in pdsw, mRpL4, membrane of the mitochondrion (reviewed in Pfanner and mRpL17 also exhibit the same phenotypes as mu- and Geissler 2001). tants isolated from our screen (Mandal et al. 2005). The generation of ATP in eukaryotic cells is largely This is an independent confirmation that this screen dependent on oxidative phosphorylation in mitochon- is uniquely capable of selecting for disruptions in cell dria (Figure 5). ATP originates from the oxidation of cycle regulation in the developing eye disc, caused by NADH and FADH2 via the electron transport chain, the activation of a metabolic checkpoint initiated by which consists of a series of four electron transporting the mitochondrion. However, not all of the mutants complexes (I–IV) and the terminal complex V, which isolated in this screen are expected to be disruptions in generates ATP. Complexes I and II transfer electrons, mitochondrial proteins. In Mandal et al. (2005), we from NADH and FADH2, respectively, to ubiquinone described a pathway linking mitochondrial processes to (CoQ) via a series of iron–sulfur clusters. Ubiquinone, cell cycle regulation in the nucleus. Two key members which is a mobile carrier, transfers these electrons to of this pathway, AMPK and p53, are known cytosolic complex III. Cytochrome c, another mobile carrier proteins. This pathway is not yet fully understood, and protein, then delivers the electrons from complex III we expect that our screen will also provide missing to complex IV where it reduces dioxygen to water. cytosolic links, in addition to mitochondrial ones de- Concurrent with electron transport, complexes I, II, scribed here. Our mapping results show that the screen and IV pump protons from the matrix into the inter- is very likely to have identified some nonmitochondrial membrane space, forming an electrochemical proton genes. For example, JV212 maps to 98E1-98E3 (Table 1, gradient across the inner membrane, which is used to A and B), a region encompassing six genes, none of generate ATP by ATP synthase or complex V. Five mu- which bears homology to other mitochondrial genes or 532 T. S. Vivian Liao et al.

Figure 5.—Schematic of complexes involved in oxidative phosphorylation in the mitochondrion. Proteins highlighted in pink are encoded by mutant genes identified from this screen or described in Mandal et al. (2005). contains a mitochondrial localization sequence. This components, will result in an enforcement of the G1–S is again true for JR8 (Table 1, A and B), which maps to checkpoint of the cell cycle. The fact that each of the 100B9, another region containing five genes that are characterized mutants with this suite of phenotypes not likely to be mitochondrial. encodes a protein with a mitochondrial function sug- An additional advantage to our screen is that it may gests that our method can serve as a general screen for be a more accurate predictor of mitochondrial function identifying nuclear-encoded genes with mitochondrial than the currently available in silico databases. For function. This study represents yet another example of example, MitoProt predicts relative low probabilities the power of Drosophila genetics and mutant screens of mitochondrial localization for CCHL (28%), Pdsw in the discovery and refinement of important cellular (10%), GatA (56%), and ATL (8.4%). We have demon- processes. strated through immunostaining in this study that both We thank Julia Thompson, Kha Nguyen, and Ryan Skophammer for GatA and ATL localize to the mitochondrion. Pdsw, a their help with the screen and the Bloomington Stock Center and the NADH dehydrogenase, and CCHL have been shown Exelixis Stock Center for fly stocks. This study was supported by in previous functional studies to play roles in complexes National Institutes of Health grant R01-EY08152 to U.B. U.B. is a I and IV, respectively, of the electron transport chain. Howard Hughes Medical Institute (HHMI) professor and G.C. is an HHMI instructor. We acknowledge HHMI for supporting research Although MitoProt will be able to reliably predict efforts by our undergraduate students, six of whom (A.C., J.M., K.Y., mitochondrial localization for the proteins that fit its Q.A.F., G.L.G., and W.K.) are included here as authors. algorithm, our screen provides the more relevant in vivo information. This study identified several new mutations in genes LITERATURE CITED that function in either oxidative phosphorylation or Ackerman, S. H., and A. Tzagoloff, 2005 Function, structure, and mitochondrial protein biosynthesis. Consistent with our biogenesis of mitochondrial ATP synthase. Prog. Nucleic Acid earlier observation using tend (Mandal et al. 2005), Res. Mol. Biol. 80: 95–133. Altschul, S. F., W. Gish,W.Miller,E.W.Myers and D. J. Lipman, these new mutants exhibit normal early cell divisions, 1990 Basic local alignment search tool. J. Mol. Biol. 215: 403– aG1–S block at the third larval instar, and proper 410. neuronal differentiation. The new mutants help to Ashby, M. N., S. Y. Kutsunai,S.Ackerman,A.Tzagoloff and P. A. Edwards, 1992 COQ2 is a candidate for the structural gene further establish that attenuated mitochondrial func- encoding para-hydroxybenzoate:polyprenyltransferase. J. Biol. tion, caused by a loss in one of its many individual Chem. 267: 4128–4136. Note 533

Boyer, P. D., 1997 The ATP synthase: a splendid molecular ma- Pfanner, N., and A. Geissler, 2001 Versatility of the mitochondrial chine. Annu. Rev. Biochem. 66: 717–749. protein import machinery. Nat. Rev. Mol. Cell Biol. 2: 339–349. Claros, M. G., and P. Vincens, 1996 Computational method to Ready, D. F., T. E. Hanson and S. Benzer, 1976 Development of the predict mitochondrially imported proteins and their targeting Drosophila retina, a neurocrystalline lattice. Dev. Biol. 53: 217– sequences. Eur. J. Biochem. 241: 779–786. 240. Diekert, K., G. Kispal,B.Guiard and R. Lill, 1999 An internal Schon, A., C. G. Kannangara,S.Gough and D. Soll, 1988 Protein targeting signal directing proteins into the mitochondrial inter- biosynthesis in organelles requires misaminoacylation of tRNA. membrane space. Proc. Natl. Acad. Sci. USA 96: 11752–11757. Nature 331: 187–190. Drysdale, R. A., and M. A. Crosby, 2005 FlyBase: genes and gene Spradling, A. C., D. Stern,A.Beaton,E.J.Rhem,T.Laverty et al., models. Nucleic Acids Res. 33: D390–D395. 1999 The Berkeley Drosophila Genome Project gene disrup- Dumont, M. E., T. S. Cardillo,M.K.Hayes and F. Sherman, tion project: single P-element insertions mutating 25% of vital 1991 Role of cytochrome c heme lyase in mitochondrial import Drosophila genes. Genetics 153: 135–177. and accumulation of cytochrome c in Saccharomyces cerevisiae. Sylvester, J. E., N. Fischel-Ghodsian,E.B.Mougey and T. W. Mol. Cell. Biol. 11: 5487–5496. O’Brien, 2004 Mitochondrial ribosomal proteins: candidate Dunn, S. D., D. T. McLachlin and M. Revington, 2000 The second genes for mitochondrial disease. Genet. Med. 6: 73–80. stalk of Escherichia coli ATP synthase. Biochim. Biophys. Acta Thibault, S. T., M. A. Singer,W.Y.Miyazaki,B.Milash,N.A. 1458: 356–363. Dompe et al., 2004 A complementary transposon tool kit for Flores, G. V., A. Daga,H.R.Kalhor and U. Banerjee, 1998 Loz- Drosophila melanogaster using P and piggyBac. Nat. Genet. enge is expressed in pluripotent precursor cells and patterns 36: 283–287. multiple cell types in the Drosophila eye through the control of Tzagoloff, A., and A. Shtanko, 1995 Mitochondrial and cyto- cell-specific transcription factors. Development 125: 3681–3687. plasmic isoleucyl-, glutamyl- and arginyl-tRNA synthetases of Flores, G. V., H. Duan,H.Yan,R.Nagaraj,W.Fu et al., 2000 Com- yeast are encoded by separate genes. Eur. J. Biochem. 230: binatorial signaling in the specification of unique cell fates. Cell 582–586. 103: 75–85. Weber, J., A. Muharemagic,S.Wilke-Mounts and A. E. Senior, Graack,H.R.,andB.Wittmann-Liebold, 1998 Mitochondrial ribo- 2004 Analysis of sequence determinants of F1Fo-ATP synthase somal proteins (MRPs) of yeast. Biochem. J. 329(Pt 3): 433–448. in the N-terminal region of alpha subunit for binding of delta Green, D. E., 1966 Comprehensive Biochemistry, pp. 309–326. Elsevier, subunit. J. Biol. Chem. 279: 25673–25679. Amsterdam. Wilcox, M., and M. Nirenberg, 1968 Transfer RNA as a cofactor Kranz, R., R. Lill,B.Goldman,G.Bonnard and S. Merchant, coupling amino acid synthesis with that of protein. Proc. Natl. 1998 Molecular mechanisms of cytochrome c biogenesis: three Acad. Sci. USA 61: 229–236. distinct systems. Mol. Microbiol. 29: 383–396. Winrow, M. J., and H. Rudney, 1969 The incorporation of Mandal, S., P. Guptan,E.Owusu-Ansah and U. Banerjee, p-hydroxybenzoic acid and isopentenyl pyrophphate into ubiqui- 2005 Mitochondrial regulation of cell cycle progression during none precursors by cell-free preparations of rat tissues. Biochem. development as revealed by the tenured mutation in Drosophila. Biophys. Res. Commun. 37: 833–840. Dev. Cell 9: 843–854. Wolff, T., and D. F. Ready, 1991 The beginning of pattern forma- Newmeyer, D. D., and S. Ferguson-Miller, 2003 Mitochondria: re- tion in the Drosophila compound eye: the morphogenetic furrow leasing power for life and unleashing the machineries of death. and the second mitotic wave. Development 113: 841–850. Cell 112: 481–490. Yao, K. M., M. L. Samson,R.Reeves and K. White, 1993 Gene elav Newsome, T. B., B. Asling and B. J. Dickson, 2000 Analysis of of Drosophila melanogaster: a prototype for neuronal-specific Drosophila photoreceptor axon guidance in eye-specific mosaics. RNA binding protein gene family that is conserved in flies and Development 127: 851–860. humans. J. Neurobiol. 24: 723–739. Ohta, S., 2003 A multi-functional organelle mitochondrion is in- volved in cell death, proliferation and disease. Curr. Med. Chem. 10: 2485–2494. Communicating editor: T. C. Kaufman